Characteristics of Transient Elastography for Hepatic Steatosis and Fibrosis in Children Undergoing Liver Biopsy.

Abstract

To determine the diagnostic performance of vibration-controlled transient elastography (VCTE) measured the same day as liver biopsy in screening and risk stratification of steatosis and fibrosis in children and adolescents. We compared VCTE measures of hepatic steatosis and fibrosis with liver biopsy histology as the reference standard. Sequential patients undergoing liver biopsy in a hospital-based pediatric gastroenterology practice were included. In primary analyses, we determined area under the receiver operating characteristic (AUC), sensitivity, and specificity for VCTE-derived measures of steatosis and fibrosis. In post hoc subgroup analyses, we examined VCTE measures of steatosis and fibrosis among patients according to metabolic dysfunction-associated steatotic liver disease (MASLD) diagnosis. Among 145 children, with mean age 13.6 years (SD: 5.1 years), VCTE measures of any steatosis (≥5%) showed AUC 0.89 (95% CI 0.83-0.94), sensitivity 0.90 (95% CI 0.79-0.97), and specificity 0.74 (95% CI 0.64-0.83). VCTE measures for fibrosis stage F2 or greater had AUC 0.71 (0.59-0.82), sensitivity 0.75 (0.53-0.90), and specificity 0.63 (0.54-0.72). In post hoc subgroup analyses among the subgroup of 37 patients with MASLD, VCTE had low ability to detect greater grades of steatosis (AUC 0.61 for S2 or greater) and low ability to detect clinically meaningful fibrosis (AUC 0.56 for F2 or greater). In the overall cohort, VCTE measurements on the same day as liver biopsy showed moderate ability to detect steatosis and limited ability to detect F2 fibrosis. Diagnostic discrimination for steatosis and fibrosis was low among the subgroup with MASLD. Disease-specific studies, rather than heterogeneous cohorts of mixed disease types, are needed to advance clinical implementation of noninvasive steatosis and fibrosis measures.